hi guys,
i am new in this forum, i have a diy projector
using sharp 1650 ,elmo hp4kdx, and viewsonic vb50hrtv
still not statisfy, i want to try build my own from 250watt ,
i 've search the net for big fresnel lens, i found this link, i am not
sure if this one will work,
here is the link
http://www.alltronics.com/lenses.htm
i am new in this forum, i have a diy projector
using sharp 1650 ,elmo hp4kdx, and viewsonic vb50hrtv
still not statisfy, i want to try build my own from 250watt ,
i 've search the net for big fresnel lens, i found this link, i am not
sure if this one will work,
here is the link
http://www.alltronics.com/lenses.htm
Well now I'm confused. Proto, your gonna have to get detailed here. Are there rings visible on your image? If so, how far away is each fresnel from the LCD?
I'm leaning towards thinking jcbklyny (I had to go back 3 times to spell that right) is right, because doesn't the projection lens focus on the upper fresnel? That would mean that there will be rings no matter what. For me personally, that is no good. But please, let us know if moving the top fresnel away from the LCD makes a difference.
I'm sure that the light distribution is better, but I had my LCD butted against my fresnel for a while and the rings drove me insane. It looks much better now w/o them.
As far a the problems with the light going in at an angle, what exactly are the problems? This is not the cause of hotspots or brighter areas at all. What are the negative effects? Is it those lines you see on moving images (mostly on fleshtones)? Would an LCD with a better viewing angle solve this?
I'm leaning towards thinking jcbklyny (I had to go back 3 times to spell that right) is right, because doesn't the projection lens focus on the upper fresnel? That would mean that there will be rings no matter what. For me personally, that is no good. But please, let us know if moving the top fresnel away from the LCD makes a difference.
I'm sure that the light distribution is better, but I had my LCD butted against my fresnel for a while and the rings drove me insane. It looks much better now w/o them.
As far a the problems with the light going in at an angle, what exactly are the problems? This is not the cause of hotspots or brighter areas at all. What are the negative effects? Is it those lines you see on moving images (mostly on fleshtones)? Would an LCD with a better viewing angle solve this?
Lifter said:
No, an LCD with faster on/off times is only thing that could solve this ghosting. Also reducing the refresh rate could help. With 16 ms (4+12ms) like the BenQ 567s theorethical max. refresh rate is 62.5 FPS, though it can update the picture at much faster rate. It would be nice to have 50 FPS rate for PAL, though HTPC can convert the stream to PAL60 easily. With 25 ms reaction time you can get only 40 Hz refresh rate, it should be OK if youre player could output just at 25 or 30 FPS rate. I don't know if there are player software that could use one refresh rate but lower FPS rate, practically that would mean that with 60 Hz refresh rate every other frame should be dropped (shouldn't be too difficult in software, though it may be that no MPEG2/DVD codecs have this feature).
When I said proto5's looked better, I was restating what other diy people said when he posted. Many of them also said his1 5{ er, but I do understand he crappy digicam thing. Mine is ok, but sucks in some situations and a person can have a good projector and the camera not do it justice. 🙂
Like said in the above post, response time is lcd's is a critical thing to look at. I haven't seen the NEC 1545v in action, but the 30 ms response time does make me nervous about it. For a good lcd around 20 ms is about right. That is good for games and the likes, though the manufacturers specs aren't always right because different manufacturers measure their specs differently. Anybody ever went to http://www.tomshardware.com ? Great site and they really put lcd's through their paces. If you want to know the best lcd for a diy project, this is the site to go to.
Like said in the above post, response time is lcd's is a critical thing to look at. I haven't seen the NEC 1545v in action, but the 30 ms response time does make me nervous about it. For a good lcd around 20 ms is about right. That is good for games and the likes, though the manufacturers specs aren't always right because different manufacturers measure their specs differently. Anybody ever went to http://www.tomshardware.com ? Great site and they really put lcd's through their paces. If you want to know the best lcd for a diy project, this is the site to go to.
The responce time on the 1545v is a little scary i will admit, but a few people have use i and said it looks great for games and what not. I own one and am using it, it look great. I was play CS on it and what not.
moegaloman said:
I must say those are some big *** fresnel lenses. Depending on the specs of those fresnels, we might be able to **** them down to size. Finally a newbie who does research! 🙂
edit haha where you see the **** a spelled cut with an extra n. 🙄
new fresnel info
While digging through my Edmund Optics catalog I came across this tidbit:
"Fresnel lenses are a compromise between efficiency and image quality: Low groove density allows HIGHER quality images, while high groove density yields better efficiency- (such as LIGHT GATHERING applications)"
Very interesting. It seems the forum's consensus is backwards. Shame we cant use a high groove count under the LCD for max light gathering and a low groove count above the LCD for best image. But that would make rings and funny moire patterns, wouldn't it?
No guts, no glory...I'll have to try it.
WTH..it worked on the split fresnels!
While digging through my Edmund Optics catalog I came across this tidbit:
"Fresnel lenses are a compromise between efficiency and image quality: Low groove density allows HIGHER quality images, while high groove density yields better efficiency- (such as LIGHT GATHERING applications)"
Very interesting. It seems the forum's consensus is backwards. Shame we cant use a high groove count under the LCD for max light gathering and a low groove count above the LCD for best image. But that would make rings and funny moire patterns, wouldn't it?
No guts, no glory...I'll have to try it.
WTH..it worked on the split fresnels!
verbose mustafa said:
no good. Shipping for those fresnels is $40. Would would be better off having customs made.
verbose: keep your mind on fresnels! 🙂
don't forget that lens had a 40" focal length, im not sure if that is what we want for this project.
check this link out
http://www.gci-net.com/users/e/ekemp/arcade/misc.htm
here is a link on that page
http://www.gci-net.com/users/e/ekemp/arcade/fresnel_lens.JPG
that wouldn't seem like too bad a setup to me, if the lens is good for this project.
Just head to a tv repair shop, no shipping
we could all be running 24" LCDs!
don't forget that lens had a 40" focal length, im not sure if that is what we want for this project.
check this link out
http://www.gci-net.com/users/e/ekemp/arcade/misc.htm
here is a link on that page
http://www.gci-net.com/users/e/ekemp/arcade/fresnel_lens.JPG
that wouldn't seem like too bad a setup to me, if the lens is good for this project.
Just head to a tv repair shop, no shipping
we could all be running 24" LCDs!
Proto5,
Your research is interesting. It would make sense from a manufacturing standpoint that courser grooves could yield better results because the smaller and tighter the grooves are the less accurate the face of the groove can be made to reflect the ideal curvature of the lens.
I wonder what the optimum line count would be, but I'm sure that the quality of the lens would be of paramount importance.
The thing is that only the light that hits the second (condensing) fresnel at a perpendicular or very close to perpendicular angle gets focused. Any light the hits at too much of an angle gets refracted in another direction. I wonder if a courser line count would make this problem worse becase there would be more surface area for non perpendicular light to hit the sides of the prisms instead of the face and be refracted back in another direction.
There must be something like an ideal compromise. The ideal thing would be for the sides of the prisms to be painted black or something.
Hezz
Your research is interesting. It would make sense from a manufacturing standpoint that courser grooves could yield better results because the smaller and tighter the grooves are the less accurate the face of the groove can be made to reflect the ideal curvature of the lens.
I wonder what the optimum line count would be, but I'm sure that the quality of the lens would be of paramount importance.
The thing is that only the light that hits the second (condensing) fresnel at a perpendicular or very close to perpendicular angle gets focused. Any light the hits at too much of an angle gets refracted in another direction. I wonder if a courser line count would make this problem worse becase there would be more surface area for non perpendicular light to hit the sides of the prisms instead of the face and be refracted back in another direction.
There must be something like an ideal compromise. The ideal thing would be for the sides of the prisms to be painted black or something.
Hezz
Verbose,
there are a couple of free high end lens design software pakages that you can download. One is called OSLO LT and the other something like Synopsis. The learning curve is relatively high for both and I am hoping that later this summer I might have some time to find out what a necessary plano or double convex lens would have to be sized to work for a 14 inch focal length and 15 inch monitor.
My hunch is that the curvature would be quite large and the lens difficult to make. But if it could work it might solve all of the fresnel issues. Of course large singlets have problems of thier own.
If you have the time and want to try these programs out see if you can design a singlet lens.
Hezz
there are a couple of free high end lens design software pakages that you can download. One is called OSLO LT and the other something like Synopsis. The learning curve is relatively high for both and I am hoping that later this summer I might have some time to find out what a necessary plano or double convex lens would have to be sized to work for a 14 inch focal length and 15 inch monitor.
My hunch is that the curvature would be quite large and the lens difficult to make. But if it could work it might solve all of the fresnel issues. Of course large singlets have problems of thier own.
If you have the time and want to try these programs out see if you can design a singlet lens.
Hezz
Psionic,
The 40 inch focal length is too long to work with nearly all OHP setups. You could build your own projector but you would need a very large objective lens or else one with a very long focal length. The focusing cone of the fresnel needs to be inside the inner diameter of the objective lens when the objective is at the correct focusing distance from the LCD. Bear in mind that the fresnel should be about an inch above the LCD so that the objective is not focusing directly on the fresnels lines.
Hezz
The 40 inch focal length is too long to work with nearly all OHP setups. You could build your own projector but you would need a very large objective lens or else one with a very long focal length. The focusing cone of the fresnel needs to be inside the inner diameter of the objective lens when the objective is at the correct focusing distance from the LCD. Bear in mind that the fresnel should be about an inch above the LCD so that the objective is not focusing directly on the fresnels lines.
Hezz
bad info-sorry
Hezz:
After checking Edmund Optics web page and finding an exact OPPOSITE statement, I called Fresnel Tech & spoke with their tech guy. He said higher groove count = better image, lower groove count = better efficiency. He also said different groove counts could be mixed, as long as the fresnels are not touching.
Hezz:
After checking Edmund Optics web page and finding an exact OPPOSITE statement, I called Fresnel Tech & spoke with their tech guy. He said higher groove count = better image, lower groove count = better efficiency. He also said different groove counts could be mixed, as long as the fresnels are not touching.
if you guys are woried about the the rings in the frensel and want the lcd with even light well why not try a bit of perspex 3mm thick lightly sandblasted? u could put it after the frensel u know after frensel before the lcd, and u could change it or experiment with the idea on the courseness of the holes from the sandblast. I think if u wer to use small or big holes as in the sand grit used u could make a screen that would diffuse the light more evenly, hell it might even make a good projection screen lol anyway thats my idea, it kinda works like the glass in the back of a backlight but for light going strait through nopt at an angle, my eyes are sore from playing from my light, i think ill go to my room and cry now lol
Trev
Trev
just curious to see how much we intensify our light, what if we wer to magnify in a way to intensify the element in the bulb before the condenser? i recon u could get a gain of 25%
Trev
still blind and crying
Trev
still blind and crying
Hi guys,
This is just for your information. Below is a screen shot from OSLO LT of a large singlet lens that I tried to design in this software package. It is available for free download. The program is quite sophisticated and I hope that this simple lens is right.
This is a 15" round lens with the first convex side a 28" radius and the second more curved convex side a 12.5" radius. The lens is 3.03" thick and has a focal length of about 13.94".
My guess is this is about what would be required for a normal singlet lens to replace the fresnel. It could be built to high standards by hand from glass or optical grade plastic but would probably cost more to build than a custom fresnel.
Hezz
P.S. The picture wouldn't post. Too large.
This is just for your information. Below is a screen shot from OSLO LT of a large singlet lens that I tried to design in this software package. It is available for free download. The program is quite sophisticated and I hope that this simple lens is right.
This is a 15" round lens with the first convex side a 28" radius and the second more curved convex side a 12.5" radius. The lens is 3.03" thick and has a focal length of about 13.94".
My guess is this is about what would be required for a normal singlet lens to replace the fresnel. It could be built to high standards by hand from glass or optical grade plastic but would probably cost more to build than a custom fresnel.
Hezz
P.S. The picture wouldn't post. Too large.
Guys,
I was wrong on the thickness. It is about 3.77" at the center of the lens. A lot of material to shine through and an expensive piece of plastic to buy. The only way that it would be economical to build is if it were cast from optical grade acylic pellets such as are used to make eyeglass lenses. The only problem there is that I don't know of a hand grinding technique to get the lens perfectly spherical unless it is slowly ground from a two flat blanks.
That brings up another expense. The tool which is an identical piece of plastic which is used to grind the lens must be thick enough to take on the contour depth of the curve. That means buying a second blank of plastic at least 3" thick or deep enough for the deep convex side. The other side is not as deep and would require a blank of about 1" - 1 1/2" thickness.
It might be cheaper to find a company that has the machines to custom grind a very large lens. Especially if they could make it aspheric for less distortion.
You can see why fresnels are popular.
Hezz
I was wrong on the thickness. It is about 3.77" at the center of the lens. A lot of material to shine through and an expensive piece of plastic to buy. The only way that it would be economical to build is if it were cast from optical grade acylic pellets such as are used to make eyeglass lenses. The only problem there is that I don't know of a hand grinding technique to get the lens perfectly spherical unless it is slowly ground from a two flat blanks.
That brings up another expense. The tool which is an identical piece of plastic which is used to grind the lens must be thick enough to take on the contour depth of the curve. That means buying a second blank of plastic at least 3" thick or deep enough for the deep convex side. The other side is not as deep and would require a blank of about 1" - 1 1/2" thickness.
It might be cheaper to find a company that has the machines to custom grind a very large lens. Especially if they could make it aspheric for less distortion.
You can see why fresnels are popular.
Hezz
Hezz,
Making a lense is a bigger project than I would take on, but sonds very interesting. If you are set on making your own, you might want to look at:
http://www.efn.org/~mbartels/tm/largthin.html
You might be able to adapt the processes he uses to make his large telescope mirrors to produce your lense.
Regards,
BT
Making a lense is a bigger project than I would take on, but sonds very interesting. If you are set on making your own, you might want to look at:
http://www.efn.org/~mbartels/tm/largthin.html
You might be able to adapt the processes he uses to make his large telescope mirrors to produce your lense.
Regards,
BT
- Status
- Not open for further replies.